The current application proposes follow-up of a sample for a funded Genome-wide Association Study (GWAS) of alcohol dependence (1799 alcohol dependent cases and 1808 controls). This cohort, which has a demonstrated commitment to research participation of more than a decade, is unique in that is comprised of individuals who were ascertained from prior community-based (rather than clinical) assessment and who grew up during a period of very restrictive Australian divorce laws so that 85% of those reporting a history of parental alcoholism were raised by both biological parents through age 16. The project's aims are: AIM1 To assess lifetime history of severe childhood and adult environmental stressors, DSM-IV lifetime diagnoses of PTSD and illicit drug dependence, and updated histories of alcohol consumption measures and DSM-IV alcohol dependence in this sample of Australians for whom whole genome SNP genotyping is in progress AIM2 Using a quantitative heaviness of drinking factor score, to test for moderation of SNP genotype effects by history of severe childhood stressors AIMS To test for moderation of SNP genotype effects on alcohol dependence symptoms by history of severe childhood stressors The current application draws on data from the project director's recently-completed Trauma Study (AA13446) which comprehensively assessed severe childhood stressors in a partially-overlapping sample of twins and siblings. We present data demonstrating the feasibility (section 3.2) of the assessment (on which the current proposal's assessment is based) including strong evidence for its utility in collecting retrospective history of childhood events (i.e excellent within-pair agreement for same-sex twins), and the ability to derive factors from its use that have excellent psychometric properties (section 3.3), display a dose-response realtionship with report of parental alcohol problems (section 3.4), and which contribute significantly to the proposal's primary dependent measure, a quantitative alcohol consumption factor (section 3.5). We believe that the proposed design is uniquely well-suited to uncover G x E interactions contributing to the liability for alcohol consumption and has adequate power for this purpose (section 4.5).